A neural network framework for similarity-based prognostics
| Autor: | Shankar Sankararaman, Indranil Roychoudhury, Jeffrey Alun Jones, Kai Goebel, Oguz Bektas |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Computer science
Clinical Biochemistry Complex system 010501 environmental sciences Similarity based RUL calculation Machine learning computer.software_genre Data-driven prognostics 01 natural sciences 03 medical and health sciences Engineering Similarity (psychology) lcsh:Science ComputingMethodologies_COMPUTERGRAPHICS 030304 developmental biology 0105 earth and related environmental sciences Protocol (science) 0303 health sciences Artificial neural networks Artificial neural network business.industry Health condition Condition monitoring Health indicator Medical Laboratory Technology Prognostics lcsh:Q Artificial intelligence business computer |
| Zdroj: | MethodsX MethodsX, Vol 6, Iss, Pp 383-390 (2019) |
| ISSN: | 2215-0161 |
| DOI: | 10.1016/j.mex.2019.02.015 |
| Popis: | Graphical abstract Highlights • The proposed method can provide a multi regime normalization process for noisy degradation trajectories. • With the neural network library, it is possible to calculate health indicators of different trajectories with distinct wear levels. • The similarity based remaining useful life estimation method can increase the prognostic performance for even short-term test subsets with little information. Prognostic performance is associated with accurately estimating remaining useful life. Difficulty in accurate prognostic applications can be tackled by processing raw sensor readings into more meaningful and comprehensive health condition indicators that will then provide performance information for remaining useful life estimations. To that end, typically, multiple tasks on data pre-processing and predictions have to be carried out such that tasks can be assessed using different methodological aspects. However, incompatible methods may result in poor performance and consequently lead to undesirable error rates. The present research evaluates data training and prediction stages. A data-driven prognostic method based on a feed-forward neural network framework is first defined to calculate the performance of a complex system. Then, the health indicators are used in a similarity based remaining useful life estimation method. This framework presents a conceptual prognostic protocol that overcomes challenges presented by multi-regime condition monitoring data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|